Beyond the B-cell


Do we have the right cell target in MS? Yes and no; we need a multicellular approach.

Recently the attention in MS has been on the B-cell as if it was the holy grail of MS treatments. It is not.

In several posts, over the last few weeks, I have made the case that the B-cell is important, probably as an antigen presenting cell, but it is not the ‘be all and end all’ of MS treatments. It is clear that rebound post-natalizumab is driven my B-cells and the positive data on the first BTK inhibitor would indicate that the B-cells are working via the B-cell receptor on antigen presentation. If only we knew what these antigens were we would have a much better handle on the cause of MS.

I know this science stuff is hard, but it is important. At the end of the day, the nut and bolts of MS must be molecular; molecules mean treatment targets and potentially more focused and hopefully better and safer treatments in the future.

I have stressed that simply targeting B-cells in both the periphery and central nervous system will not be enough to effectively treat MS in the long-term. When we look at end-organ damage markers in pwMS who are on B-cell therapies they have ongoing brain volume loss, albeit at a lower rate, and enlarging lesions (T1 black holes), which are both indicative of ongoing smouldering MS. So what do we need to do? I have provided circumstantial evidence that NIRTs (non-selective immune reconstitution therapies) have a slight edge on the B-cell therapies and this may be because they are also targeting T-cells. The latter, however, comes at a price of greater adverse events in relation to immunosuppression. The proportion of MSers on NIRTs who experience disability improvement seems higher when compared to the anti-B cell agents, which indicates that NIRTs are doing something else over and above their effect on the B-cell compartment. However, based on their overall safety profile it is unlikely that the NIRTs (alemtuzumab & HSCT) will be a therapeutic strategy that the wider MS community will adopt with vigour. Although from comments on this blog there is an informed group of MSers who feel hard done by because their HCPs won’t offer them the option of using NIRTs first-line, i.e. very early in the course of their disease when they have the most to gain from these therapies.

Is there anything else we can do to improve on the profile of B-cell therapies to make them better? Yes, I think there is. Targeting the plasma cell,in addition to the B-cell. Data on plasma cells goes back decades and surprisingly the plasma cell has never been a major therapeutic target in MS. John Prineas, one of my MS heroes, has always stressed the importance of the plasma cell in MS. His paper below from 1978 documents just how enriched the brains of MSers are with this population of cells. What is often not stressed is that the biology of plasma cells is so so different to the B-cell, which opens up new therapeutic targets that are quite different to those in the T and B cell compartments. More on this topic another time.

You are aware of the recent publication showing that about 55% of Polish MSers treated with intravenous cladribine lost their oligoclonal bands 10 or more years after treatment and if they did lose their OCBs they tended to have lower EDSS scores. We have known for years that MSers, with either relapse-onset or primary progressive diseases, who don’t have OCBs do better. There is also evidence from biomarker and pathology studies that the OCBs may be driving several of the disease processes that have been linked to advanced or progressive MS, i.e. microglial activation and grey matter pathology. Based on these observations, we hypothesise that OCBs are very likely to be pathogenic in MS, which is why we are embarking a research programme to try and target the plasma cells within the CNS of MSers. Do you think we are crazy?

To get a handle on the plasma cell we are going to have to study what happens in the spinal fluid. There are simply too many plasma cells in the periphery which will drown out any signal from the CNS. To participate in the studies we are planning we will have to perform serial, annual, lumbar punctures or spinal taps to see if our add-on therapy is killing and/or reducing the number of plasma cells in your brain and spinal cords. The good news is that we have de-risked the lumbar puncture with the use of atraumatic needles and screening. I never thought I would be saying this but most of our patients don’t mind having LPs, particularly when they understand the reason behind the LP.  CSF neurofilament levels are now part of our prognostic profile of MSers at baseline and we are increasingly using them to assess response, or lack of response, to treatment. So if you want to be treated and treated-2-target beyond NEDA, and beyond the B-cell, then having an LP is important.

We hope our proposed plasma cells studies will lead to a mindset that goes beyond the B-cell to target some of the mechanisms that are responsible for smouldering MS.

Prineas & Wright. Macrophages, lymphocytes, and plasma cells in the perivascular compartment in chronic multiple sclerosis. Lab Invest. 1978 Apr;38(4):409-21.

Perivascular cells in CNS tissue from six multiple sclerosis (MS) patients and a patient with motor neuron disease were examined by light and electron microscopy. Lymph node tissue from one MS patient was also examined. CNS perivascular macrophages in both MA and motor neuron disease were found to closely resemble free macrophages elsewhere in the body except that they often contained unusually large primary lysosomes. Cytoplasmic inclusions consisting of membrane-bound stacks of curved linear profiles, presumed to be a product of myelin degradation, were constantly observed in microglia in MS plaques but were rarely observed in perivascular macrophages in the same area. Unidentified cylindrical bodies were observed within cysternae of rough endoplasmic reticulum in some lymph node cells. Quantitative studies of the perivascular cell population in one MS case revealed, in histologically normal white matter 260 lymphocytes and 178 plasma cells per cubic millimeter of fresh tissue. Typical chronic plaque tissue without obvious inflammatory cell cuffing contained 1772 plasma cells per cubic millimeter of fresh tissue. No plasma cells were observed in the CNS in motor neuron disease. The results of this study suggest that perivascular macrophages in the CNS represent a specialized population of monocyte-derived free macrophages, that these cells differ functionally from microglial cells, and that the digestion of myelin breakdown products in MS requires the participation of both cell types. The results also suggest that in some chronic MS cases there is a large, permanent population of CNS plasma cells that persists, like the elevated cerebrospinal fluid IgG level in this disease, for the life of the patient, that these cells, rather than inflammatory cells in fresh lesions, are the major source of this raised IgG, and that the existence of such a population of cells may indicate the continuing expression of antigens in chronic MS lesions in the absence of fresh lesion formation.

About the author

Prof G

Professor of Neurology, Barts & The London. MS & Preventive Neurology thinker, blogger, runner, vegetable gardener, husband, father, cook and wine & food lover.


  • Any hint on what drugs you will be testing in these studies? Can you please post more details on these studies? I am interested in participating. Thank you.

  • Excellent post. Very thought provoking. Do you really think all the people with MS who are on rituximab or ocrelizumab will come back with SPMS?

      • Not a nice thing to hear, but the devil you know I guess. Excellent post, this blog is truly excellent! How do you explain the apparent improved efficacy of B cell depleting agents in patients with Gd enhancing lesions at baseline?

        • Re: “How do you explain the apparent improved efficacy of B cell depleting agents in patients with Gd-enhancing lesions at baseline?”

          This is a class effect and is seen with all DMTs and is not unique to B-cell depleting agents. Gd-enhancement identifies a more active population in terms of focal inflammation and hence more likely to have relapses and/or progressions and hence more likely to be informative in clinical trials. I must stress that even in the cohorts of patients without Gd-enhancing lesions we see a treatment response.

    • Because OCBs are part of an immune response. MS is not the only immune-mediated disease to affect the CNS. We see OCBs in infections and paraneoplastic diseases.

      • Are you convinced that OCBs are pathogenic?

        So how to understand this:

        Type I IFNs, which include IFN-β, elevate expression of B cell activation factor (BAFF), increase B cell activity and drive the production of autoantibody in systemic lupus erythematosus (SLE) and neuromyelitis optica (NMO), promoting inflammation(1–3). In
        one sense, these are “type 1 IFN diseases” where B cell autoantibody production is clearly pathogenic. In RRMS IFN-β also increases serum levels of BAFF and B cell activity(4, 5), yet in a seeming paradox IFN-β reduces inflammation and decreases relapses(6).

        Interferon-β treatment requires B cells for efficacy in neuro-autoimmunity.1

        J Immunol. 2015 March 1; 194(5): 2110–2116. doi:10.4049/jimmunol.1402029.

  • Cladribine gets into the CNS but remind me does it target predominantly B cells which are not the cells which seem to get the best results for MS

    • Cladribine targets both B-cells and T-cells, albeit to a lesser extent. It may have an effect on early plasma cell lineages, but based on the expression data of the metabolic enzymes responsible for its mode of action, cladribine is unlikely to have a major effect on plasma cells. However, the empirical data on CSF OCBs from the Scripps trial and now the Polish data suggests it may have, which is why we are actively investigating the biology of cladribine on CNS B-cells and plasma cells.

  • “You are aware of the recent publication showing that about 55% of Polish MSers treated with intravenous cladribine lost their oligoclonal bands 10 or more years after treatment”

    How do they know that OCBs didn’t disappear sooner than after 10y? If I understood correctly, they didn’t do serial LPs each year, just after 10y. So although this is really important to know, it does bring even more interesting questions. I perceive Cladribine as more than a SIRT, as it deplete T cells too. Maybe we just need to adjust dosing not by weight, but also by how much B’s T’s and memory cells it depletes?

    As you said that you do a lot of LPs now in your clinic, how many patients loose their OCBs? It would be interesting to know these numbers for other DMTs as well. You’re sitting on the trove of data with , just dig in 🙂 It doesn’t even cost a lot. And if you need help with this, I’m sure there is a lot of academics reading this blog who could and would help you.

    • The findings of this Polish study needs to be confirmed; one swallow does not make a summer. We had several small case series published suggesting MSers lost OCBs on natalizumab, but a large more definitive German multicentre study showed this was not the case.

      In infectious diseases such as neurosyphilis when the condition is treated with antibiotics the OCBs are still there after 10 years. A subset of plasma cells is long-lived, which is why we need to target them with specific therapies.

      Regarding loss of OCBs in our clinical practice; this is something we can look into. But we have only started doing LPs for NFL in the lost 3 years and then in quite a selected patient cohort. We will get back to you on this. But we have another study specifically looking into this.

  • Thanks for post Dr. G. Sorry, a few questions:

    If B-cell therapy, like Ocrevus, has no or little effect on BVL or grey matter loss on MRI which is associated with disease progression, then why is Ocrevus approved as a treatment for progressive patients? It has a barely statistically significant < 25% decline on progression of disease in a selection biased trials choosing only Gd+ progressive MS patients. I guess same could be said for Tysabri, both a B- and T-cell inhibitor across the BBB, which has no effect on grey matter loss as well in a recent publication. I would imagine Gilenya would have the same outcome if trials in progressive patients were not aborted. Is there any correlation between eliminating oligoclonal banding and halting progression of disease? Stroke patients, amongst other CNS diseases, have these non-specific oligoclonal bands as well. Don't you think oligoclonal banding just indicates damage to CNS and maybe a red herring as a successful endpoint in treatment of MS? Wouldn't decreasing NFL correlate better with halting disease progression than eliminating oligoclonal banding? What proof is there that B-cells are the APC in MS? Why is it not hot microglia or anything else in the "field"?

    • I think you have misinterpreted my posts. B cell therapies have a very good effect on disease activity and render a large proportion of pwMS NEDA-3. However, they don’t appear to normalise rates of brain volume loss or close down the slowly expanding lesions. In other words, we need add-on treatments over and above B-cell therapies to target these other mechanisms. All I am saying is that we must not be lulled into a false sense of security that if you are NEDA on B-cell therapies or for that matter any therapy that you are okay and never going to get worse. Many of my patients who are NEDA-3 get worse, which is why I did a long blog post on this topic. What we are proposing are add-on studies to go beyond what B-cell therapies offer. This is clearly needed.

    • Re: “What proof is there that B-cells are the APC in MS? Why is it not hot microglia or anything else in the “field”?”

      Circumstantial proof based on (1) speed of the onset of action of anti-CD20 therapies, which is too quick to be due an impact on antibody production, (2) the positive trial results of a BTK inhibitor that prevents B-cell receptor signalling and (3) the fact that an anti-CD20 prevents rebound post-natalizumab. Yes, I am sure other cells could be APCs and they may be driving a lower-levels of inflammation, which I hypothesise in my post on smouldering MS.

      • Correct me if I am wrong but the B-cell receptor appears to be more related activating B cells, not presenting antigen to T cells

        • It is the immunoglobulin receptor, which picks up antigen and internalises it. The surface immunoglobulin makes the B-cell a very effective APC as it allows very small amounts of antigen to be detected and internalised. Antigen presentation via the B cell is very different to other APCs.

    • Re: “…Why is Ocrevus approved as a treatment for progressive patients?”

      Because the trial was positive, i.e. it reduced disease progression by ~25% compared to placebo. However, it did not stop progression, which is why we need other therapies over and above ocrelizumab to tackle the myriad of disease processes in so-called non-relapsing progressive MS.

  • Part of the problem with patients fears of lumbar punctures might be branding. Noone in their right mind would want their lumbar “puncturing”. Come up with a softer name for the procedure.

  • Isn’t this phrase contradicting itself (as “Almost everyone who has relapse onset MS has SPMS already” and MS is one disease etc…)?
    “We have known for years that MSers, with either relapse-onset
    or primary progressive diseases, who don’t have OCBs do better. ”
    If you can develop PPMS without OCBs, it means that eliminating OCBs will not stop progression. So you are targeting at something vague and modest.
    ( I would really love to see data about this claim).

    • Most people with MS are losing brain from the earliest stages of the disease; even in the asymptomatic stages of the disease. If you define progressive MS as the chronic loss of neurones and axons then this phase of the disease is there from the beginning. Depressing I know, but if you wait for the clinically apparent progressive phase to appear then it is rather late to make a big difference to the long-term trajectory of this disease.

      • Thank you, but it doesn’t answer to my question:
        “If you can develop PPMS without OCBs, it means that eliminating OCBs will not stop progression. So you are targeting at something vague and modest.”

        • I don’t think OCB negative MS exists; I predict when we find the cause of OCB-ve MS will be another disease. For our study, you have to have OCBs to participate. We did the same of the ocrelizumab in PPMS or ORATORIO study.

          • So, who is this who does better?A non MS patient?

            I really hope it is just me who sees condradtictions in your theory about OCBs.

    • There is a big difference between a blog post and a research article. May be you should ask Prof G to stop blogging and spend his time doing research 😉

      • I was actually comparing the two John Prineas articles :/ ProfG is doing a strategic work in this blog and it is irreplaceable, imo..

  • What an extremely informative, accessible blog this is. I am in my first year following diagnosis with fairly active MS based on MRI and symptoms, recently started on a B cell therapy. Obviously the information here makes me question if I should self advocate for moving up to a non-specific immunomodulator early on. My neurologist seemed hesitant to track csf neurofilament levels via periodic lp to gauge my responsiveness to ocrelizumab. Is CSF monitoring an unreasonable request? It seems like evidence supports basing decisions on more than MRI activity. Input on this issue would be welcome.

  • Since all relapsing patients progress to SPMS, can we still consider that alemtuzumab and HSCT a potential cure?

    Or are they merely jut slowing the disease progression, very much like natalizumab?

    • Not sure you are correct in assuming all people become SPMS; unless you call ageing SPMS. The experiment is running; are those patients who are NEDA-4 post-HSCT and post-alemtuzumab 10+ years later cured or not? An interesting question which I have been urging Genzyme to explore in a deep phenotyping study for some time.

      • NEDA 4 ! I thought everybody suffered from end-organ damage.

        Are we saying that some HSCT and alemtuzumab are NEDA 4 after 10 years, but not a single natalizumab patient is?

  • Is the role of Plasma Cells (PC) positive or negative?

    “Removal of plasmablast (PB) plus PC resulted in exacerbated EAE that was normalized by the introduction of gut-derived IgA+ PC. Furthermore, mice with an over-abundance of IgA+ PB and/or PC were specifically resistant to the effector stage of EAE, and expression of interleukin (IL)-10 by PB plus PC was necessary and sufficient to confer resistance.”

By Prof G



Recent Posts

Recent Comments